Platform Updates March 2026: GPT-5.4, Mercury 2, Gemini 3.1 Flash Lite, and More

Since our last blog update in February 2026, Palabres has continued expanding its model catalog and shipping performance improvements to make every conversation faster and smoother.

New AI Models Added Since February 2026

OpenAI updates

• Added GPT-5.4 — OpenAI's latest generation flagship model
• Added GPT-5.4 mini — a mid-tier reasoning model balancing speed and capability
• Added GPT-5.4 nano — the smallest and fastest model in the GPT-5.4 family, ideal for quick tasks
• Added GPT-OSS 20B and GPT-OSS 120B — OpenAI's open-weight models, served at cost via Cloudflare Workers AI on the free tier

Google updates

• Added Gemini 3.1 Flash Lite (preview) — Google's ultra-fast, lightweight model for high-volume, low-latency use cases

Inception Labs updates

• Added Mercury 2 — the second generation of Inception's high-speed diffusion language model, with improved quality over Mercury 1

MiniMax updates

• Added MiniMax M2.7 — an updated version of MiniMax's frontier model with improved instruction following

Z.ai / Zhipu AI updates

• Added GLM-5 Turbo — a faster and more efficient variant of the GLM-5 model family

Platform and Performance Improvements

Beyond model additions, we shipped several improvements to make the day-to-day experience better:

• Improved first message experience — conversations now start more fluidly with a smoother initial interaction
• Optimized conversation history — faster loading and rendering of past conversations
• General performance optimizations — reduced latency across the board for returning users

Try it out

All of this is live now.

Explore all models | Start a new chat | View pricing | Contact us